Over the past year or so, PaaS – Platform as a Service - has emerged as a new and interesting strategy for Salesforce.com with Force.com and NetSuite with NS-BOS. The value proposition for developers is:
- Object-oriented development environment for rapid prototyping and application development.
- “Out-of-the-cloud’ integration with other applications developed on the platform
- No need to invest in commodity operational infrastructure such as “ping, power, and pipe” and disaster recovery services which can be expensive to set up yet completely non-differentiating.
So, if you’re a SaaS application developer or investor, what’s not to like?
Well, first let’s talk about the SaaS business model.
One of the key drivers for this model to generate profits similar to traditional enterprise software companies is the ability to amortize the cost of the service over your customer base. As your base grows, the infrastucture costs/customer or COGS/customer must be able to decline to a point where they represent just a few percentage points of gross margin.
If those variable costs remain relatively fixed at a high percentage of each user, then the SaaS model – which is already a difficult one to reach profitability quickly – is weakened substantially. I suspect the PaaS providers will ameliorate that with a discounted cost as your user counts grow.
So, while a critical issue, this isn’t the primary concern I have with PaaS providers who also sell their own applications. The real concern I have is the potential downstream conflict of interest.
What do I mean?
Let’s say you’re a developer and you see a big opportunity to build a solution for widget management. So, you build a widget management application using the tools/platform from your favorite PaaS vendor. Initially, everything is just great — you’re up and running quickly and you don’t have to hire a bunch of ops people to run your 24/7 widget managemet application software application.
You quickly begin to grow because your widget management application solves a legitimate business problem. At some point, your PaaS provider – who knows everything about your installed base because they manage your infrastructure – becomes aware of the growth you’re achieving in the widget management market and decides that they, too, want to enter the this market. “Overnight” they are no longer your “friend/partner”, they are now your chief competitor.
And, unlike Microsoft, which both partners and competes with its development community, your PaaS vendor knows everything about your customers – who they are, usage rates, adoption rates, etc. And, they can decide to “work less cooperatively” with you. You no longer get return phone calls to your support issues. You don’t get active cooperation with the product organization. Etc, etc.
Now you are really stuck. You’ve built your application using their development environment and their platform. What do you do? Your only option is to rewrite the application and get it moved onto another PaaS provider that doesn’t also sell applications — as fast as possible. This could be a very onerous undertaking and potentially not survivable.
So, unless you’re a company that is going to build applications for small markets, I would have to think long and hard about tying myself so closely to a company that might end up being my biggest competitor.
I would want to understand the warrantees, the SLA, the financial remuneration policies available to me before I embarked down this path. I’m sure this isn’t what Salesforce or NetSuite want to hear but it is definitely an issue that needs to be addressed head on if they want to achieve their PaaS objectives.